Consolidation of the Financial Services Industry: Implications for Credit Unions

Transcription

1 Consolidation of the Financial Services Industry: Implications for Credit Unions A Colloquium at Stanford University Research Contributors: Discussants: Allen N. Berger Rebecca S. Demsetz Timothy Guinnane Philip E. Strahan Linda L. Baughman Henry W. Wirz Coloquium Sponsors: Center for Credit Union Research, University of Wisconsin-Madison Filene Research Institute Graduate School of Business, Stanford University Stanford Federal Credit Union Prepared for the Center for Credit Union Research and the Filene Research Institute P.O. Box 2998 Madison, Wisconsin (608)

2 Copyright 1999 by Filene Research Institute. ISBN All rights reserved. Printed in U.S.A.

3 Filene Research Institute and Center for Credit Union Research The Filene Research Institute is a non-profit organization dedicated to scientific and thoughtful analysis about issues affecting the future of consumer finance and credit unions. It supports research efforts that will ultimately enhance the wellbeing of consumers and will assist credit unions in adapting to rapidly changing economic, legal, and social environments. Deeply imbedded in the credit union tradition is an ongoing search for better ways to understand and serve credit union members and the general public. Credit unions, like other democratic institutions, make great progress when they welcome and carefully consider high-quality research, new perspectives, and innovative, sometimes controversial, proposals. Open inquiry, the free flow of ideas, and debate are essential parts of the true democratic process. In this spirit, the Filene Research Institute grants researchers considerable latitude in their studies of highpriority consumer finance issues and encourages them to candidly communicate their findings and recommendations. The name of the institute honors Edward A. Filene, the father of the U.S. credit union movement. He was an innovative leader who relied on insightful research and analysis when encouraging credit union development. The Center for Credit Union Research is an independent academic research center located in the School of Business at the University of Wisconsin Madison. The Center conducts research and evaluates academic research proposals on subjects determined to be priority issues by the Research Council of the Filene Research Institute. The Center also supervises Filene Research Institute projects at other universities and institutions. The purpose of the Center s research is to provide independent analysis of key issues faced by the credit union movement, thus assisting credit unions and public policymakers in their long-term planning. Progress is the constant replacing of the best there is with something still better! Edward A. Filene i

4 ii

5 Table of Contents Executive Summary and Overview by Harold O. Fried and William A. Kelly, Jr. Chapter 1: Financial Consolidation: Implications for Small Financial Services Firms and Their Customers by Allen N. Berger, Rebecca S. Demsetz and Philip E. Strahan Introduction A Decade of Financial Consolidation The Causes of Financial Consolidation Why Has Consolidation Accelerated? The Consequences of Financial Consolidation The Consequences of Consolidation for Small Customers Conclusions and Implications References Chapter 2: Discussion of Financial Consolidation in the U.S Discussant Comments by Linda L. Baughman Excerpts from Question and Answer Session Chapter 3: Consolidation of Financial Services in Europe: The Response of German and French Credit Unions by Timothy Guinnane Introduction Banking Systems In Germany and France Credit Unions In Germany Credit Unions In France The Cooperative Niche In Germany and France Conclusions: Implications for U.S. Credit Unions Selected References Chapter 4: Discussion of Financial Consolidation in Europe Discussant Comments by Henry W. Wirz Excerpts from Question and Answer Session iii

6 About the Authors and Discussants Participants at the Colloquium Filene Research Institute Administrative Board/ Research Council Filene Research Institute Publications iv

7 Executive Summary and Overview Harold O. Fried William A. Kelly, Jr. INTRODUCTION The effect of mega-banks on credit unions has become an increasingly compelling issue as financial services consolidation continues to accelerate. To address this issue, the Filene Research Institute; the Center for Credit Union Research, University of Wisconsin-Madison; Graduate School of Business, Stanford University; and Stanford Federal Credit Union jointly sponsored a colloquium, held at Stanford University in March This monograph contains two key papers that were presented at the colloquium, as well as selected portions of the discussion of participants. Those participants included a mixture of credit union CEOs and academics. At a colloquium, every participant takes away his or her own views based on the papers presented and the ensuing discussion. In this overview, we present our own view of the key points presented in the papers and discussions, along with our view of the most important implications for credit unions. Others may not agree with all of these points, but we hope the ideas will serve as a basis for healthy discussion and more informed strategic thinking about how credit unions can adapt to a rapidly changing landscape financial services landscape. CONSOLIDATION TRENDS The first paper in this volume, Financial Consolidation: Implications for Small Financial Services Firms and their Customers, summarizes key consolidation trends in the financial services industry since 1988 and the underlying causes for these changes. Some key findings are: Deregulation seems to have caused less consolidation than one might expect. For example, California has allowed branch banking since Therefore, national deregulation would be expected to have little effect on consolidation there. Yet the degree of consolidation that took place among small banks in California during the period 1988 to 1997 is similar to that which took place in the U.S. as a whole during this period. This means that U.S. credit unions should not expect consolidation to decrease as deregulation nears completion. It appears that new 1

8 changes in the marketplace, such as technological innovation, rather than deregulation, may be primary causes of consolidation. Although the number of bank charters and the number of bank holding companies has declined since 1988, the number of bank offices has risen. While this data does not measure the degree to which these offices are full service, it does contradict the perception that consolidation has reduced the total number of branch offices available to customers. Consolidation has increased banking concentration in the country as a whole. For example, the largest eight banking institutions had 22.3% of deposits nationally in 1988, increasing to 35.5% in However, households usually buy their financial services in local markets. Metropolitan Statistical Areas (MSA s), on average, showed a slight decline in concentration during the same period, as did counties outside MSA s. We see two important implications from this relatively constant level of bank concentration in local markets. First, despite consolidation, consumers have benefited from essentially the same level of competition among banks in their local market, even though more banks may have non-local headquarters. Second, despite consolidation, the competition credit unions face in their local markets has not decreased in terms of the size and number of institutions. However, more competing institutions are branches of large banking institutions headquartered outside the local area. Competition from these institutions is characterized by both the advantages and disadvantages which the branches of large banks exhibit toward consumers. Consolidation has reduced the national market share of banks with less than $300 million in assets, from 12.6% in 1988 to 9.9% in This suggests small banks compete less effectively in the marketplace than larger banks. Banks under $50 million in assets fared even less well. During the same period, their national market share dropped from 8.4% to 5.9%. During the same period, credit union assets increased from $197 billion to $360 billion. However, the percentage of assets held in credit unions with under $300 million in assets fell from 83% to 2

9 63%, and the percentage held in credit unions under $50 million dropped fell from 38% to 19%. 1 Therefore, even though credit unions serve primarily consumers and banks serve primarily commercial customers, smaller institutions as defined above have clearly lost ground during this period. REASONS FOR CONSOLIDATION Why has consolidation taken place? Dozens of research studies have examined the reasons for bank consolidation. Many studies find considerably less evidence of economies of scale than one might expect from looking at changes in the market share of smaller institutions. Researchers have employed a variety of statistical approaches and data periods without reaching a clear consensus on the question of optimal size. Some estimates find little evidence of operational economies in banks beyond sizes as small as $100 million. Other studies provide evidence for economies of scale up to $1 billion, and a few even higher. However, mergers producing $50 billion banks and larger are puzzling in light of the existing state of research on economies of scale in banking. The implication is that either the very largest banks are merging past the point of gaining economies of scale, perhaps even making themselves less efficient, or researchers have not yet hit upon reasons why these mergers occur. For now, the implication for credit unions is that the cost advantages of very large banking competitors, as reported in much of the trade and popular press, could be exaggerated. If that s the case, then branches of very large banks may not offer more favorable rates to consumers than credit unions offer. If this is true, mega-banks may be able to out-compete credit unions only on the number or quality of services offered. This suggests that competing with mega-banks on services offered may be more important than competing on price, where bank advantages could be exaggerated. Empirical evidence over a number of years indicates that credit unions consistently outperform banks in consumer satisfaction surveys. This suggests that credit unions excel at high quality service, where their local roots and relatively small size may be an 1 Source: CUNA Economics and Statistics 3

10 advantage. However, credit unions operating individually could be at a disadvantage in service offerings that require economies of scale. This implies the following key strategy for credit unions: to innovate in ways that allow them to be the initial point of contact with the member, while a source much larger than the credit union provides as much as possible of the remainder of such services. At the same time, credit unions should retain and even enhance the key roles they play in offering services that can be more effectively produced at the level of the individual credit union, such as lending. LOCAL BANK COMPETITION Recent studies on economies of scale in banking provide significant implications for credit unions strategic outlook. These studies have found significant economies of scale for banks that practice geographical diversification of their loan portfolios. Although consumer loan defaults are reasonably stable in the face of local economic declines, banks face more risks because their loans are typically to local businesses. Therefore, by merging across states, they can diversify their loan portfolio more effectively. This doesn t change the average return on their portfolio. However, it does reduce the fluctuation around the average. They need a lower capital ratio to guard against fluctuations of a portfolio with the same return. Since bank capital is primarily the equity of stockholders, geographical diversification means banks can get the same average return with less equity invested; hence stockholders receive a higher return on their equity. To make this happen, banks acquire existing smaller institutions in diverse geographical areas. It doesn t mean that the banks acquired were poorly managed or had other problems, or that the acquiring bank could necessarily run the local bank any better than the existing owners. The implications for credit unions are: First, if a large, out-of-state bank buys a local bank, it does not necessarily mean that the local bank will be run any more effectively or more competitively than it was before. It may even be run less effectively. Second, as out-of-state ownership creates a void in service to commercial customers who prefer local decision-making, a 4

11 newly chartered bank may well enter the market. During the period 1988 to 1997, when rapid consolidation occurred, new banks entered the market at a median rate of 148 per year, compared to a median rate of 171 cross-market mergers per year. This explains in part why local market concentration did not rise over this period while national concentration did. EXPERIENCES ABROAD Much of the banking consolidation that has taken place in recent years in the U.S. was experienced long ago in Europe. Credit unions have existed in Europe since the middle of the nineteenth century. Therefore, it is instructive to know what happened to credit unions there during banking consolidation. This is especially important because the credit union model that was implemented in the U.S. came from Europe via Canada, so the original organization of credit unions here resembled closely that of European credit unions at the time. U.S. and European credit unions share common roots. The fundamental, defining characteristics of U.S. and European credit unions remain unchanged. That is, they are not only owned, but also effectively controlled, by members. One-member, onevote democratic procedures elect members of the board. However, European credit unions have adapted to competition against mega-banks differently than U.S. credit unions have. And European credit unions have thrived. As described in the second paper here, Consolidation of Financial Services Industries In Europe: The Response of German and French Credit Unions, European credit unions have competed successfully with mega-banks by using a three-tiered delivery system. This system allows local credit unions to deliver a wide variety of services, with gains from economies of scale exploited at the regional and national levels. The system is more thoroughly developed than in the United States. Germany and France differ in the degree to which local credit unions are committed to the system. In Germany the commitments are by choice, although in practice almost all credit unions participate. In France, commitment is required and credit unions have less local autonomy. However, in both systems, the use of a common name within a given national system is used to give a greater national 5

12 presence. In Germany there is more than one national group, and local institutions have more flexibility about local identification within the national system. A few other key differences developed as European credit unions competed with mega-banks. Local credit unions in Europe are larger than in the U.S. For example, German credit unions average about $235 million in assets. They have innovated in how they build capital, so that they do not depend entirely on retained earnings. They can serve anyone they wish and, along with banks, they can offer insurance and securities services. They also pay taxes within the European tax system. With these changes from common roots they have surpassed the U.S. in some respects and lag a bit in others. They hold nearly a fifth of net bank deposits in Germany and about a quarter in France. However, their membership penetration in the population is somewhat less than the U.S. For example, the 14 million German members are a smaller fraction of its population than the 76 million members in the U.S. This could reflect the fact that the European credit unions provide commercial as well as consumer services. CONCLUSIONS A common perception of how credit unions must respond to the U.S. trend toward mega-banking can be expressed as grow or die. Our view, based on the research presented and the related discussion at the Stanford colloquium, is different. While the growth doctrine has an element of truth, it is far from the whole truth. A more accurate implication is to exploit the economies of scale where they benefit the member. This can be done through innovative support arrangements for smaller institutions as well as through very large institutions that work alone. Further, the efficiencies of large and small depend on what types of support arrangements are available or created. The European experience provides a tangible and thriving example of using such a support system in competing with megabanks, without local credit unions growing to the size of large banks. The European experience also provides examples of innovation in raising capital and broadening service offerings. 6

13 In light of these findings, one participant at the colloquium suggested that a more accurate perception of how U.S. credit unions must respond to mega-banks could be expressed as, merge or cooperate. U.S. credit unions can choose merger, and develop into very large institutions working alone, or they can choose cooperation and innovation in order to develop as relatively smaller institutions working together. We see clear opportunities for credit unions to operate at a variety of successful sizes using local roots, personal service and cooperation. The option of merging into very large organizations that work alone is the traditional choice of banks. Studies of competitive strategies show that those, which attempt to avoid competition, are ineffective. A better strategy is to embrace competition as a dynamic, changing presence, but to maintain an advantage by competing from a position of strength. 7

14 8

15 CHAPTER 1: Financial Consolidation: Implications for Small Financial Services Firms and their Customers Allen N. Berger Rebecca S. Demsetz Philip E. Strahan (The opinions expressed in this chapter do not necessarily reflect those of the Federal Reserve Board, the New York Reserve Bank or their staffs. This chapter draws heavily on an overview paper designed to explore the causes, consequences and policy implications of financial consolidations by Berger, Demsetz and Strahan, 1999.) INTRODUCTION The structure of the financial services industry in the U.S. has changed dramatically in recent years as consolidation, particularly in the banking sector, has taken off. While there has been considerable research on financial services industry consolidation, much is yet to be understood. The main purposes of this article are to bring together the most up-to-date research on the topic, and to discuss the implications of consolidation for small financial service providers and the traditional customers of these firms small depositors and small businesses. We begin by reporting the facts of financial consolidation in the U.S. over the past decade. We then outline a framework to understand this consolidation. In particular, we argue that the main motivation behind consolidation is to maximize shareholder value, although we also consider the motives of other stakeholders, particularly managers and governments. Value may be maximized through mergers and acquisitions (M&As) primarily by increasing the participating firms market power in setting prices, by improving their efficiency, and in some cases by increasing their access to the safety net. Our framework predicts that the pace of consolidation will be determined primarily by changes in economic environments that alter the constraints faced by financial service firms. We discuss four such changes that may be partially responsible for the recent rapid pace of consolidation technological progress, improvements in financial condition, excess capacity or financial distress in the industry or market, and deregulation of geographical or product restrictions. Next, we describe the consequences of consolidation for market power and efficiency. Our framework divides the research on the consequences of consolidation into static analyses and dynamic analyses. Static analyses relate the potential consequences of consolidation to certain characteristics of financial institutions associated with consolidation, such as size. While static studies do not use data on M&As and are not necessarily intended to provide information about the effects of consolidation, they may be useful in predicting the consequences of M&As. For example, static analyses of scale efficiency may give valuable information on the efficiency effects of M&As in which the institutions substantially increase their size. Dynamic analyses 9

16 compare the behavior of financial institutions before and after M&As or compare the behavior of recently consolidated institutions with other institutions that have not recently engaged in M&As. Dynamic analyses take into account that M&As are dynamic events that may involve changes in organizational focus or managerial behavior. These analyses also incorporate any short-term costs of consummating the M&A (legal expenses, consultant fees, severance pay, etc.) or disruptions due to downsizing, meshing of corporate cultures, or turf battles. Dynamic studies can also account for the external effects of consolidation, defined here as the reactions of other financial service providers to M&As in their markets. The literature suggests that potential for gains in market power may provide a motivation for consolidation. However, measures of local-market concentration have changed little over the past decade, suggesting that market power has not been greatly affected. By contrast, the efficiency effects of consolidation have been very important. Consolidation seems to permit increases in scale, scope and profit efficiency, and to help diversify the portfolio risks of the participants on average. This finding is particularly strong in more recent studies that have used data from the 1990s, suggesting that technological changes in financial services may have increased economies of scale and scope. The likely increase in the efficient scale of production in banking may be reducing the role of small banks and, along with deregulation, helps explain their declining market share. While there is much less research on the scale and scope economies of nonbank financial institutions, these results suggest that other small financial services firms such as thrifts and credit unions may also experience a declining share as a result of consolidation. We also review the research literature on the potential consequences of consolidation for the availability of services to small customers, which have traditionally been disproportionately served in large part by small financial service firms. The research suggests that the effects of consolidation on the availability of services to small customers are likely to be small. Large banking organizations generally devote fewer of their assets to loans to small businesses, and M&As involving large banks are generally found to reduce small business lending by the participants. When small banks merge, however, small business lending appears to 10

17 increase. In addition, limited evidence suggests that other institutions in the local market may make up most of the lost credit supply. M&As may also reduce service availability to small depositors through branch office closings, although the limited evidence suggests these closures occur infrequently and generally occur only when there is another branch office nearby. A DECADE OF FINANCIAL CONSOLIDATION Tables 1-4 report aggregate statistics on trends in financial consolidation in the U.S. Structural changes in the U.S. banking industry from M&As, de novo entry, and failure are shown in Table 1. Several hundred M&As occurred each year, about half of the in-market type and half of the market-extension type. During this period, megamergers M&As between institutions with assets over $1 billion each became common, most of them occurring between institutions in different states. Although not shown in the table, some very recent M&As in the U.S. and elsewhere have increased dramatically in size, some reaching the scale of supermegamergers M&As between institutions with assets over $100 billion each. Based on market values, nine of the ten largest M&As in U.S. history in any industry occurred during 1998, and four of these Citicorp-Travelers, BankAmerica- NationsBank, Banc One-First Chicago and Norwest-Wells Fargo occurred in banking (Moore and Siems, 1998). In addition, the recent UBS-Swiss Bank Corp. supermegamerger created the largest bank in Europe. 11

18 Table 1 M&As, Failures, and Entry in the Commercial Banking Industry Total M&As of Unaffiliated Banking Organizations In-Market Market Extension NA NA Mega- Mergers [Interstate] 14 [7] 3 [2] 5 [2] 16 [12] 23 [15] 15 [10] 15 [11] 20 [16] 26 [14] 15 [11] Other changes in industry structure occurred as high rates of failure reduced the number of banks while high rates of de novo entry increased the number of banks. During the latter part of the 1980s, each year about 200 banks failed and about 200 new banks were formed, although the de novo banks were generally much smaller than the failed banks. Both failure and entry declined in the early 1990s, while rates of entry picked up in the mid-1990s as industry profitability rose. De Novo Entry Failures Source: Rhoades (1996, Tables 4, 10, 18) and Meyer (1998). Figures for 1997 are estimated. Mergers are defined here as combinations that bring together under common ownership banks that had been independent. Mergers of affiliated banks are not included. When a multibank holding company is acquired, each commercial bank in that holding company is counted as a separate merger. Market extension mergers are defined as those where the target bank is located in a different market from the acquirer, where a market is defined as a Metropolitan Statistical Area (MSA) or non-msa county. Both the target banking organization and the acquiring banking organization must have total assets over $1 billion to be considered a megamerger. 12

19 As a result of this consolidation and restructuring activity, the number of U.S. banks and banking organizations stand-alone banks and top-tier bank holding companies (BHCs) both fell by almost 30% between 1988 and 1997 (Table 2). During this period, the share of total nationwide assets held by the largest eight banking organizations rose from 22.3% to 35.5%. We also look at measures of concentration at the local level defined as a Metropolitan Statistical Area (MSA) or non-msa county given the evidence discussed below that markets for most retail banking products are local. Despite consolidation activity, the average local market deposit Herfindahl index actually declined slightly over the period, falling by about 4% for MSAs and by about 5% for non-msa counties. Total bank offices rose by 16.8%, although total bank plus thrift offices declined by 0.1% in part because banks acquired branches formerly owned by failed thrifts. Number of U.S. Bank Charters Table 2 Concentration, Ownership, and Number of Firms in the Commercial Banking Industry Number of Banking Organizations Number of Banking Offices Number of Offices in Banks plus Thrifts Eight Firm Concentration Ratio Deposit Herfindahl in MSAs Deposit Herfindahl in Non-MSA Counties ,130 9,881 59,518 84, % 2,020 4, ,727 9,620 60,720 84, % 2,010 4, ,370 9,391 62,753 84, % 2,010 4, ,949 9,168 63,896 83, % 1,977 4, ,496 8,873 63,401 81, % 2,023 4, ,001 8,446 63,828 80, % 1,994 4, ,491 8,018 65,597 81, % 1,976 4, ,984 7,686 66,454 81, % 1,963 4, ,575 7,421 67,318 82, % 1,991 4, ,216 7,234 69,463 83, % 1,949 4,114 Source: Reports of Income and Condition and NIC, , FDIC Historical Statistics on Banking, and Meyer (1998). A banking organization is a top-tier bank holding company or a stand-alone bank. The concentration ratio is based on total assets of domestically chartered banks. The Deposit Herfindahl is 10,000 times the sum of squared market shares based on deposits for banking organizations operating in Metropolitan Statistical Areas (MSAs) or Non-MSA counties. All figures are year-end. 13

20 The market shares of very small and small banking organizations defined as banking organizations with total assets below $50 million (1997 dollars) and banking organizations with total assets between $50 and $300 million fell sharply between 1988 and The share of domestic assets held by small banking organizations fell from 12.6% to 9.9%, and the share of assets held by very small banking organizations fell from 3.5% to 1.8% (Table 3). As we discuss in the next section, the accelerated pace of consolidation has occurred, in part, because states deregulated their restrictions on banks ability to expand both within and across state lines Table 3 Asset Market Share of Small Banking Organizations Market Share for All of U.S. Very Small Bakning Organizations Small Banking Organizations It has been argued that California provides a simple benchmark for what the U.S. banking system would look like in the absence of regulations restricting bank expansion (Berger, Kashyap and Scalise, 1995). California has a large and well-diversified economy that is bigger than the economy of most nations. In addition, branching has been permitted there since 1909, and California has permitted interstate banking since 1987 (Jayaratne and Strahan, Market Share for California Very Small Banking Organizations Small Banking Organizations Source: Reports of Income and Condition, , year-end. Banking organizations are stand-alone banks or top-tier bank holding companies. A very small banking organization is one with total banking assets below $50 million in 1997 dollars; a small banking organization is one with assets between $50 million and $300 million in 1997 dollars. Market share figures are based on domestic assets. For multi-state holding companies, only assets held by California bank subsidiaries are counted in columns 3 and 4. 14

21 1998). The share of assets held by very small and small banking organizations in California was 1.1% and 8.4% in 1988, compared to 3.5% and 12.6% overall (Table 3). The much lower shares for very small and small banking organizations in California supports the idea that bank regulation restricting expansion in other states artificially raised the market share of small banks, and deregulation may have reduced their market share over time. However, trends in the market shares of very small and small banking organizations suggest that other factors are also at work, since these share have declined in recent years by similar proportions in both California and the U.S. as a whole. Very small banking organizations lost nearly three-quarters of their market share in California between 1988 and 1997, while very small banking organizations lost about half of their market share in the U.S. as a whole. 1 The market share for small banking organizations shows the same pattern. The share of assets held by these banking organizations declined by about 30% in California, compared to a decline of about 20% overall. So, deregulation of restrictions on bank expansion cannot provide a full explanation for declines in the market share of small banking organizations. Table 4 gives data for U.S. nonbank financial institutions. The life insurance, property-liability insurance, and securities brokerage industries have changed much less than banking. Note that these segments did not face restrictions on geographical expansion like the banking segment, nor did they experience significant deregulation. The securities and life insurance segments were somewhat less concentrated in the mid-1990s than in the late 1980s, while the property-liability insurance segment was somewhat more concentrated. There was a substantial reduction in the number of thrift institutions and a corresponding increase in concentration due to the high rate of thrift failure in the late 1980s and early 1990s. The credit union industry remains very unconcentrated, presumably because of its not-for-profit status and because its members must have a common bond, such as employment at the same firm. 1 The market share figures in Table 2 are for commercial banks only. Similar trends emerge using both banks and thrifts. 15

22 Life Insurance Number of Firms , , , , , , , , , N/A Table 4 Concentration in Nonbank Segments of Financial Services Asset Share of 8 Largest Firms 41.7% 40.4% 39.0% 38.1% 37.2% 36.4% 35.3% 34.9% 34.7% N/A Property-Liability Insurance Number of Firms 0,940 1,193 1,272 1,267 1,232 1,197 1,187 1,179 1,138 N/A Asset Share of 8 Largest Firms 32.5% 32.4% 32.4% 32.2% 32.2% 31.5% 31.3% 33.7% 36.1% N/A Securities Firms Number of Firms 6,432 6,141 5,827 5,386 5,260 5,292 5,426 5,451 5,553 5,597 Capital Share of 10 Largest Firms 57.5% 61.8% 63.6% 62.1% 62.2% 63.4% 60.9% 59.3% 58.5% 55.5% THE CAUSES OF FINANCIAL CONSOLIDATION In our framework, the primary motive for consolidation is maximizing shareholder value. In the absence of capital market frictions, all actions by the firm, including consolidation activities, are geared toward maximizing the value of shares owned by existing shareholders. The preferences of other stakeholders are taken into account only insofar as they affect the value of shares through the cost of funds, supply of labor or other factors of production, or the demand for services. In practice, however, managers and government often affect consolidation decisions more directly. Savings Institutions Number of Firms 3,175 3,100 2,725 2,386 2,086 1,726 1,532 1,420 1,322 1,201 Asset Share of 8 Largest Firms 13.5% 15.0% 18.2% 19.9% 19.3% 17.7% 19.2% 21.7% 21.3% 30.6% Credit Unions Number of Firms 13,875 13,371 12,860 12,960 12,594 12,317 11,991 11,687 11,392 11,238 Asset Share of 8 Largest Firms 6.3% 6.5% 6.7% 6.8% 7.4% 7.7% 7.9% 7.9% 7.8% 8.0% Source and Notes: Life Insurance Cummins, Tennyson and Weiss (1999), National Association of Insurance Commissioners, Life-Health Annual Statement Data Tapes (Kansas City, MO, annual), and American Council of Life Insurance, Life Insurance Fact Book (Washington, DC). An insurance firm is a holding company owning multiple insurance companies, or an independent insurance company. Property-Liability Insurance: Cummins, Tennyson and Weiss (1999), A.M. Best Company, Best s Insurance Reports: Property-Casualty (Oldwick, NJ, Annual), and A.M. Best Company, Best s Aggregates and Averages: Property and Casualty (Oldwick, NJ, Annual). Securities: 1998 Securities Industry Fact Book. The number of firms is the number of NASD members. Concentration is based on the share of capital at the ten largest firms since the asset share information is not available. Savings Institutions: Thrift Financial Reports, Includes savings and loan associations and federal savings banks. Credit Unions: National Credit Union Administration. Includes both federal and state chartered credit unions. 16

23 In this section, we outline and review the literature on the value maximizing and non-value maximizing motives for financial consolidation. We then identify four types of changes in the economic environment that may have led to the recent accelerated pace of consolidation. Value-Maximizing Motives Financial service firms can maximize value in one of two main ways through consolidation by increasing their market power in setting prices, or by increasing their efficiency. It is difficult to determine the goals of M&A participants, but there is evidence consistent with the notion that some M&As are designed to increase market power. Research described below suggests that inmarket M&As that substantially increase market concentration may increase market power in setting prices on retail services. Presumably, this was an expected consequence of many M&As, and provided at least part of the motivation. Research described below also suggests that M&As may increase efficiency, consistent with the presumption that expected efficiency improvements provide part of the motivation for M&As. In addition, a number of studies have found that in a substantial proportion of M&As, a larger, more efficient institution tends to take over a smaller, less efficient institution, presumably at least in part to spread the expertise or operating policies and procedures of the more efficient institution over additional resources. In the U.S., acquiring banks appear to be more cost efficient than target banks on average (Berger and Humphrey 1992; Pilloff and Santomero 1998). Another study of U.S. banks found that acquiring banks are more profitable and have smaller nonperforming loan ratios than targets (Peristiani 1993). Simulation evidence also suggests that large X-efficiency gains are possible if the best practice banks merge and reform the practices of the least efficient banks (Savage 1991; Shaffer 1993). Case studies of U.S. bank M&As support the idea that potential efficiency gains act to motivate some M&As as well (Calomiris and Karceski 1998; Rhoades 1998). However, one study of U.S. banks found that while poorly-capitalized banks are more likely to be acquired, banks with a high degree of cost inefficiency are, ceteris paribus, less likely to be acquired without government assistance (Wheelock and Wilson 1998). 17

24 Some evidence also suggests that efficiency concerns may motivate consolidation in other segments of the financial industry. Acquirers in the U.S. life insurance industry tend to be more efficient than average, and targets tend to be financially impaired (Cummins, Tennyson, and Weiss 1999). In the credit union industry, acquirers tend to be larger than average, while targets tend to be smaller and in weaker financial condition than nontarget credit unions (Fried, Lovell and Yaisawarng 1999). M&As may also improve efficiency, as broadly defined here, if greater diversification improves the risk-expected return tradeoff. Consistent with this idea, one study found that U.S. acquiring banks bid more for targets when the consummation of the M&A would lead to significant diversification gains (Benston, Hunter and Wall 1995). Diversifying M&As may also improve efficiency in the long term through expanding the skill set of managers (Milbourn, Boot and Thakor 1999). However, studies outside of financial services suggest that diversifying M&As are generally value-reducing, and that increases in corporate focus are valueenhancing (Lang and Stulz 1994; Berger and Ofek 1995; John and Ofek 1995). In addition, although it is not exactly market power or efficiency, some institutions may try to increase the value of their access to the government s financial safety net (including deposit insurance, discount window access, payments system guarantees) through consolidation. 2 If financial market participants perceive very large organizations to be too big to fail i.e., that explicit or implicit government guarantees will protect debtholders or shareholders of these organizations there may be incentives to increase size through consolidation, lower the cost of funding, and increase the value of shares. International comparisons over a 100-year period show how changes in the structure and strength of safety net guarantees may affect financial institution risk-taking, and by extension, the motive to consolidate to increase the value of access to the safety net (Saunders and Wilson 1999). 2 Regulators may also help protect large financial institutions by encouraging other institutions to lend to or invest in a financially distressed institution. 18

25 Non Value-Maximizing Motives As noted above, stakeholders other than shareholders may have a direct effect on consolidation decisions. We consider here the roles of managers and governments, who appear to have more influence over consolidation decisions for financial institutions than for nonfinancial firms. 3 Managers may be able to pursue their own objectives in consolidation decisions, particularly in banking where corporate control may be relatively weak. Banking regulations in the U.S. weaken the corporate control market by generally allowing only other banks and BHCs to acquire a bank. The regulatory approval/disapproval process may also deter some acquirers. In addition, most U.S. banks are small and are not publicly traded. Perhaps as a result, hostile takeovers that replace management are rare in U.S. banking (Prowse 1997). However, corporate control appears to improve when intrastate and interstate banking deregulation increases the number of potential acquirers, which increases market discipline, reduces the market share of poorly run banks, and generally raises profitability (Schranz 1993; Hubbard and Palia 1995; Jayaratne and Strahan 1996,1998). One managerial objective may be empire-building. Executive compensation tends to increase with firm size, so managers may hope to achieve personal financial gains by engaging in M&As, although at least in part the higher observed compensation of the managers of larger institutions rewards greater skill and effort. To protect their firm-specific human capital, some managers may also attempt to reduce insolvency risk below the level that is in shareholders interest, perhaps by diversifying risk through M&A activity. There is evidence that banking organizations may overpay for acquisitions when corporate governance structures are not sufficiently well-designed to align managerial incentives with those of owners. For example, banks that have addressed managerial agency problems through high levels of managerial shareholdings and/or concentrated ownership experience higher (or less negative) abnormal returns when they become acquirers 3 Debtholders may play a lesser role in the consolidation decisions of financial institutions than nonfinancial firms. This is because financial institution debtholders may be protected in whole or in part by the safety net, and by government supervision which tends to block M&As that create substantial risks. 19

26 than banks that have not addressed these agency conflicts as well. In addition, abnormal returns at bidder banks are increasing in the sensitivity of the CEO s pay to the performance of the firm and to the share of outsiders on the board of directors (Allen and Ceboyan 1991; Subrahmanyam, Rangan and Rosenstein 1997; Cornett, Hovakimian, Palia and Tehranian 1998). This evidence suggests that entrenched managers with little pay sensitivity to performance or outside directors may make acquisitions that do not maximize shareholder wealth. Managerial entrenchment may also prevent some value maximizing M&As by reducing the willingness of some financial institutions to become targets of M&As. One study found that banks in which managers hold a greater share of the stock are less likely to be acquired and that this effect is much larger at banks where management leaves following an acquisition (Hadlock, Houston, and Ryngaert 1999). This is consistent with the idea that management teams with large ownership stakes can block outside acquisitions. The government plays a direct role in consolidation decisions through restricting the types of M&As permitted (e.g., limits on interstate or international M&As, or M&As between banks and other firms), and through approval/disapproval decisions for individual M&As. In part, this is to limit the government s liability and prevent exploitation of too big to fail and expansion of the safety net. Regulatory review of bank M&A applications in the U.S. attempts to prevent consolidation in which excessive increases in risk are expected. Regulators also prevent in-market M&As if the increases in concentration are expected to result in excessive increases in market power. Regulators may block M&As to promote other policy goals as well. For example, U.S. banking organizations may be prevented from making acquisitions if they do not meet the lending standards of the Community Reinvestment Act (CRA). 4 4 For a review of CRA issues, see Thomas (1998). Community advocates may reinforce this government motive by petitioning regulators to block proposed M&As on these grounds. In some cases, consolidating banks make well-publicized pledges to provide credit to small businesses and make loans in low-income neighborhoods during the regulatory approval/denial process. 20

27 In contrast to these restrictions on M&As, government may also encourage consolidation beyond the safety net incentive discussed above. During periods of financial crisis, the government may provide financial assistance or otherwise aid in the consolidation of troubled financial institutions. For example, the FDIC provided financial assistance to allow healthy banks to purchase over 1,000 insolvent U.S. banks between 1984 and In other nations, governments have acquired troubled institutions themselves. WHY HAS CONSOLIDATION ACCELERATED? Our framework predicts that the pace of consolidation will be determined primarily by changes in economic environments that alter the constraints faced by financial service firms. A relaxation of constraints may allow consolidation that increases shareholder value or make it easier for managers to pursue their own goals through consolidation. We identify four key changes that may help explain the recent fast pace of M&A activity technological progress, improvements in financial condition, excess capacity/ financial distress, and deregulation. New Technologies Technological progress may have increased scale economies in producing financial services, creating opportunities to improve efficiency and increase value through consolidation. New financial engineering tools such as derivative contracts, off-balance-sheet guarantees and risk management may be more efficiently produced by larger institutions. Some new delivery methods for depositor services, such as phone centers, ATMs, and on-line banking may also give greater economies of scale than traditional branching networks (Radecki, Wenninger, and Orlow 1997; and Mishkin and Strahan, forthcoming). Advances in payments technology may also have created scale economies in back-office operations and network economies. The advent of these new technologies, and the associated scale economies, help explain the decline in the market share of small banks documented above, even in states experiencing little deregulation (e.g. California). An important caveat is that technologies embodying scale economies may in some cases be accessed at low cost by small financial institutions. The new tools of financial engineering, advanced depositor delivery services or payments technologies 21

28 may be distributed to small financial institutions through correspondent banking systems, through franchising or outsourcing to firms specializing in the technologies, shared ownership or mandatory sharing of payments networks, etc. 5 However, some evidence discussed below suggests that scale economies in banking have increased in the 1990s, consistent with technological progress that favors larger institutions. Improved Financial Condition Recent improvements in the financial condition of institutions may be another factor behind the increase in M&As. In the U.S., bank profitability broke records in the mid-1990s. Low interest rates and high stock prices also reduced financing constraints on M&A activity, although they raised the price of target firms as well. Evidence from outside banking suggests that financial condition can affect investment, although the reason why condition matters is less clear (Fazzari, Hubbard and Petersen 1988; Froot and Stein 1991; Hoshi, Kashyap, and Sharfstein 1991; Lamont 1996; Kaplan and Zingales 1997). One possibility is that external finance is more costly than internally generated funds, so firms may invest in more positive net present value projects when they are flush with cash. It is also possible that as more free cash flow becomes available, managers may choose to acquire other firms to suit their own goals (Jensen 1986). The data suggest that M&A activity in banking appears to respond more to the low interest rate and high stock price environment of the U.S. during the mid-1990s than M&A activity in nonfinancial industries, despite the fact that stock deals are more common than cash acquisitions in banking (Esty, Narasimhan, and Tufano 1999). Removal of Excess Capacity Consolidation may also be an efficient way to eliminate excess capacity that has arisen in the consolidating firms industry or local market. When there is excess capacity, some firms may be below efficient scale, have an inefficient product mix, or be inside the efficient frontier. M&As may help solve these efficiency problems. M&As may also help remove excess capacity more efficiently than 5 Despite economies in sharing technology, some institutions may keep their own information and back-office technologies to improve their future options for expansion into different activities (Thakor 1999). 22